1. Field of the Invention
The present invention relates to a method for reproducing magneto-optical information recorded by a mark-edge recording method on a recording medium by utilizing domain wall displacement, an apparatus therefor, and a method for the produced information.
2. Related Background Art
The magneto-optical information recording-reproduction apparatus employing a magneto-optical disk is promising because of the portability, large memory capacity, information erasability, rewritableness and so forth. FIG. 1 shows an optical head of a conventional apparatus for reproducing magneto-optical memory. In FIG. 1, a divergent light flux emitted from semiconductor laser 12 is paralleled by collimator lens 13, and is converted into a parallel luminous flux having a circular cross-section by beam correction prism 14. The linear polarized light components orthogonal to each other are named P-polarized light and S-polarized light respectively in the present invention. The P-polarized light flux is linear polarized light parallel to the drawing sheet face of FIG. 1. The P-polarized light flux is introduced to polarized light beam splitter 15. The polarized light beam splitter has a property, for example, such that it allows the P-polarized light to pass through at a transmittance of 60% and reflects the P-polarized light at a reflectivity of 40%, whereas it intercepts the S-polarized light at a transmittance of 0%, and reflects the S-polarized light at a reflectivity of 100%. The P-polarized light flux having passed through polarized light beam splitter 15 is condensed by objective lens 16 to be projected as a fine light spot onto a magnetic layer of magneto-optical disk 17. To this projected light spot area, an external magnetic field is applied by magnetic head 18 to record a magnetic domain (a mark) on the magnetic layer.
The reflected light from magneto-optical disk 17 is returned through objective lens 16 to polarized light beam splitter 15. A part of the reflected light is separated and introduced to an optical reproduction system, where the introduced light flux is further separated by another light beam splitter 19. Polarized light beam splitter 19 has a property, for example, such that it allows the P-polarized light to pass through at a transmittance of 20% and reflects the P-polarized light at a reflectivity of 80%, whereas it intercepts the S-polarized light at a transmittance of 0%, and reflects the S-polarized light at a reflectivity of 100%. The one light flux separated by polarized light beam splitter 19 is introduced through condenser lens 25 to half prism 26, where it is further separated into two fluxes, one being introduced to photodetector 27, and the other being introduced through knife edge 28 to photodetector 29. Such optical control systems make error signals for auto-tracking and auto-focusing of the light spot.
The other light flux separated by polarized light beam splitter 19 is introduced to polarized light beam splitter 22 through halfwave plate 20 for changing the light flux polarization direction by 45.degree. and condenser lens 21 for condensing the light flux. In polarized beam splitter 22, the introduced light flux is separated into two fluxes. The separated light fluxes are respectively detected by photodetector 23 or 24. Polarized light beam splitter 22 has transmittance of 100% and reflectivity of 0% for the P-polarized light, and transmittance of 0% and reflectivity of 100% for S-polarized light. The differential of the signals detected by photodetector 23 and 24 is detected by a differential amplifier (not shown in the drawing) to produce a reproduced signal.
A magneto-optical medium records information by a change of vertical magnetization directions. When a linear polarized light is projected onto a magneto-optical medium having information recorded by the change of magnetization direction, the polarization direction of the reflected light turns in a dextro direction or in a levo direction depending on the magnetization direction. For example, as shown in FIG. 2, the linear polarized light of the P-axis direction turns by an angle of +.theta.k to the direction R+ by reflection for downward magnetization, and turns by an angle of -.theta.k to the direction R- by reflection for upward magnetization. In FIG. 2, axis S is an axis for S polarized light direction. Points S+ and S- on axis S are S coordinates of points R+ and R- when letting the coordinates of points R+ and R- be (P+, S+) and (P+, S-) in P-S coordination system, respectively.
With an analyzer placed in the direction shown in FIG. 2, the light passing through the analyzer comprises the component A for R+, and the component B for R-. Therefore, the information can be detected by the photodetectors as the difference in the light intensity. In the example of FIG. 1, polarized light beam splitter 22 serves as the analyzer, in the direction of +45.degree. to the P axis for the one separated light flux, and -45.degree. to the P axis for the other separated light flux. The signal component detected by photodetector 23, and that detected by photodetector 24 are opposite in phase to each other. Therefore, the reproduction signal can be obtained with less noise by detecting differentially the individual signals.
Recently, for higher recording density of the magneto-optical medium, the system for the recording of magnetic domains (marks) is changing from pit-position recording to pit-edge recording. FIGS. 3A to 3E compares the two recording systems. FIG. 3A illustrates a row of pits recorded by pit position recording. FIG. 3B illustrates the detection signal of the pit row. In the pit position recording, information is recorded at the center of the magnetic domain. Although the recording is made at the magnetic domains of approximately the same size, the magnetic domains vary in size owing to the recording sensitivity of the medium. However, the positions of the centers are fixed without fluctuation, so that the information can be recorded precisely, advantageously. Therefore, early apparatuses employed the pit position recording system.
On the other hand, FIG. 3C illustrates a pit row in the pit edge recording. FIG. 3D illustrates the detection signal therefor. FIG. 3E illustrates pit edge detection signals. The edge detection signal detects zero-cross points of the detecting signal and the slice level (zero level) as shown in FIG. 3D. In the pit edge recording, the information is recorded at the edge (boundary) of the magnetic domain. A larger amount of information can be recorded in a unit area by the pit edge recording than by the pit position recording. Since the variation of the medium has lessened with recent techniques, the magneto-optical recording system is shifting from the pit position recording to the pit edge recording.
In conventional recording systems, the recording density of the optical disk like a magneto-optical medium generally depends on the laser wavelength and the NA (numerical aperture) of the objective lens of the optical reproduction system. The diameter of the light spot depends on the laser wavelength .lambda. and the NA of the objective lens. The size of the reproducible magnetic domain is limited approximately to .lambda./2NA. Accordingly, in conventional optical disk, for higher recording density, the laser wavelength is made shorter, or the NA of the objective lens is made larger. However, improvement by the laser wavelength and the NA of the objective lens is naturally limited, so that techniques are being developed for improvement of the recording density by constitution of the recording medium and the reading system.
For example, Japanese Patent Application Laid-Open No. 6-290496 discloses a domain wall displacement reproduction system. In this system, a light spot is allowed to scan tracks on a magneto-optical medium constituted of laminated magnetic layers to transfer magnetic domains formed by vertical magnetization in a first magnetic layer onto a third magnetic layer opposing the first magnetic layer with interposition of a second magnetic layer provided for controlling the exchange coupling. Thereby, the magnetic domain recorded on the first magnetic layer is expanded, and therefrom a reproduction signal is obtained.
The domain wall displacement reproduction system is explained by reference to FIGS. 4A to 4D. FIG. 4A is a sectional view illustrating construction of a magnetic layer of a recording medium. FIG. 4B is a plan view taken from a light spot-projecting side of the magnetic layer. In the drawings, the numeral 30 indicates magnetic layers of a magneto-optical disk as a magneto-optical medium without showing the substrate, a protection layer and so forth. First magnetic layer 31 is a memory layer for recording information by magnetic domains. (The first magnetic layer 31 is hereinafter referred to as a memory layer). Second magnetic layer 32 is a switching layer for controlling the exchange coupling force between memory layer 31 and third magnetic layer 33. (The second magnetic layer is hereinafter referred to as a switching layer.) Third magnetic layer 33 is a displacement layer to which the magnetic domains are transferred from memory layer 31 by the action of switching layer 32 and by utilizing thermal distribution caused by the light spot, and enlarges the size of the magnetic domains by displacement of the magnetic wall of the transferred magnetic domain to be larger than the size of the magnetic domains recorded in memory layer 31. (The third magnetic layer 33 is hereinafter referred to as a displacement layer.) The numeral 34 indicates a light spot for reproduction. The numeral 35 indicates a track for reproduction on magneto-optical disk 30. The arrows in memory layer 31, switching layer 32 and memory layer 33 show the directions of atomic spin in the respective layers. Domain walls 36 are formed at the boundaries where the direction of the spin is reversed in the layers. The numeral 37 shows a domain wall of a magnetic domain transferred to displacement layer 33 which wall is about to be moved.
FIG. 4C shows temperature distribution caused in magneto-optical disk 30. Axis T means temperatures of this medium. Axis X means positions. The reproduction by domain wall displacement may be conducted either by one light spot or by two light spots in principle. For simplicity of the explanation, a two-light spot method is described. In FIGS. 4A and 4B, only the light spot contributing signal reproduction is shown. Another light spot (not shown in the drawing) is projected to cause temperature distribution shown in FIG. 4C. At position Xs, the temperature Ts of disk 30 is nearly the Curie temperature of switching layer 32. The shadowed area 38 in FIG. 4A is the region where the temperature is not lower than the Curie temperature of switching layer 32.
FIG. 4D shows distribution of the domain wall energy density .sigma.1 in displacement layer 33 corresponding to the temperature distribution shown in FIG. 4C. Axis .sigma. means domain wall energy densities. Axis F means forces exerted to the domain wall. Such a gradient of domain wall energy density .sigma.1 in X direction exerts force F1 as shown in FIG. 4D to the domain walls at position X in the respective layers. Force F1 serves to displace the domain walls toward lower domain wall energy portion. In displacement layer 33 where the domain wall coercivity is low and the mobility of the domain wall is high, the domain wall is displaced readily only by this force F1. However, in the region before position Xs (righthand in FIGS. 4C and 4D), the temperature of the magneto-optical disk 30 is lower than Ts, so that the domain wall in the displacement layer is fixed by exchange coupling at the same position as the corresponding domain wall of memory layer 31.
In FIG. 4A, domain wall 37 is placed at position Xs. At position Xs, the temperature of magneto-optical disk 30 rises to a temperature Ts approximate to Curie temperature of switching layer 32, and thereby the exchange coupling between displacement layer 33 and memory layer 31 is broken. Then, domain wall 37 at position Xs in displacement layer 33 will be displaced instantaneously to the region having a higher temperature and a lower domain wall energy density. Therefore, with the passage of reproducing light spot 34, the atomic spins in displacement layer 33 within the light spot are all directed to the one direction as shown in FIG. 4A. Thus, with the movement of the medium, domain wall 37 (or 36, etc.) is displaced instantaneously, and the atomic spins in the light spot are all directed to the same direction.
From the light reflected by magneto-optical disk 30, a reproducing signal is detected by differential detection in the same manner as conducted in the optical head of FIG. 1. In the domain wall displacement reproduction, the signals reproduced by the light spot have a constant amplitude independently of the magnetic domain sizes recorded in memory layer 31, because of arrangement of atomic spins in one direction, whereby the waveform interference caused by optical diffraction limit is prevented. In other words, reproduction by domain wall displacement enables reproduction of magnetic domain of sub-micron linear density, smaller than the size of resolution limit .lambda./(2NA) decided by the laser wavelength .lambda. and NA of the objective lens.
FIG. 5 shows an example of an optical head applicable to domain wall displacement reproduction by a two-light spot method. In FIG. 5, semiconductor laser 39 for record reproduction, having for example a wavelength of 780 nm, and another semiconductor laser 40 for heating, having for example a wavelength of 1.3 .mu.m, are both placed so as to project P-polarized light to a recording medium. The laser beams emitted from semiconductor lasers 39 and 40 are made nearly circular by beam-forming means not shown in the drawing, and converted to a parallel light fluxes by collimator lenses 41 and 42. Dichroic mirror 43 is designed to transmit light of 780 nm with 100% transmittance, and to reflect light of 1.3 .mu.m with 100% reflectivity. Polarized light splitter 44 is designed to transmit P-polarized light with transmittance of 70 to 80%, but to reflect S-polarized light perpendicular to P-polarized light at reflectivity of about 100%.
The parallel light fluxes emitted from collimator lenses 41 and 42 are introduced through dichroic mirror 43 and polarized light splitter 44 to objective lens 45. The light flux of 780 nm is adjusted to be larger than the size of the aperture of the objective lens 45, and the light flux of 1.3 .mu.m is adjusted to be smaller than the size of the aperture of objective lens 45. Therefore, the same objective lens 45 has a smaller NA of the lens for the light flux of 1.3 .mu.m, and the size of the light spot thereof on recording medium 46 is larger than that of 780 .mu.m. The reflected light from the recording medium is changed again to a parallel light flux, and reflected by polarized beam splitter 44 to form light flux 47. From light flux 47, servo error signals or information reproduction signals are derived after wavelength separation or a like treatment to light flux 47 by optical systems not shown in the drawing in the same manner as in a conventional system.
FIGS. 6A and 6B show the relation between the record-reproducing light spot and the heating light spot on the recording medium. In FIG. 6A, the numeral 48 indicates a record-reproducing light spot having wavelength of 780 .mu.m; 49, a heating light spot having wavelength of 1.3 .mu.m; 50, a domain wall of a magnetic domain recorded on land 51; 52, a groove; and 53, an area of higher temperature formed by heating light spot 49. Record-reproducing light spot 48 and heating light spot 49 overlap with each other on land 51 formed between grooves 52. Thereby, temperature gradient is formed on the moving recording medium as shown in FIG. 6B. The temperature gradient and record-reproducing light spot 48 are in the same relation as in FIGS. 4A to 4D. Thereby, information reproduction can be practiced by utilizing the domain wall displacement. As described above, the reproduction by the domain wall displacement may be conducted by a one-beam method or a two-beam method. The one-beam method is preferred in view of simplicity of the apparatus, but has disadvantages mentioned below.
FIGS. 7A and 7B show the principle of reproduction by domain wall displacement. FIG. 7A is a sectional view of the magnetic layer of magneto-optical disk 30 similar to that in FIG. 4A. The magnetic layer is constituted of memory layer 31, switching layer 32, and displacement layer 33. FIG. 7B is a plan view taken from the light spot-introducing side, showing land 51 of the track, grooves 52, and light spot 54 for reproduction. The projection of light spot 54 causes temperature distribution as shown by egg-shaped contour lines. The medium is moved in the direction shown by arrow mark A. Arrow marks in the magnetic layers of magneto-optical disk 30 show the directions of atomic spins. Region 57 is at a temperature higher than the Curie temperature of switching layer 32, thereby switching layer 32 in that region being demagnetized. As the result, at the higher temperature region 57, the exchange-coupling between memory layer 31 and displacement layer 33 is broken without transfer of the magnetic domains in memory layer 31 to displacement layer 33. Except for this higher temperature region 57, the magnetic domains of memory layer 31 are transferred to displacement layer 33 under action of exchange-coupling force.
When domain walls 55 and 56 of recorded magnetic domains have reached the boundaries of higher temperature region 57, the domain wall 55 moves in the direction shown by arrow mark B toward the higher temperature region, and domain wall 56 moves in the direction shown by arrow mark C toward the higher temperature region. Domain wall 55 moves to region 58 (hereinafter referred to as a front region), and domain wall 56 moves to region 59 (hereinafter referred to as a rear region). As shown in FIG. 7B, a conventional differential detection method reproduces a mixture of the information of domain wall 55 and domain wall 56 without giving the desired information.
This problem, described in FIGS. 7A and 7B, is discussed in more detail by reference to FIGS. 8A to 8G. FIGS. 8A to 8F show scanning of land 51 on the track by reproducing light spot 54. The medium is being driven in a direction shown by arrow mark A similarly as in FIGS. 7A and 7B with formation of front region 58 and rear region 59. Isolated magnetic domain (mark) 60 is recorded on land 51. For example, the case is considered where only the isolated magnetic domain 60 is magnetized upward, and the other magnetic domains are magnetized downward. The numerals 61 and 62 indicate respectively the front domain wall and the back domain wall. FIG. 8G shows reproduced waveforms of a differential signal at the respective positions.
FIG. 8A shows the state of light spot 54 apart from isolated magnetic domain 60. In this state, both front region 58 and rear region 59 are magnetized downward, and the differential detection signal is at the base level as shown in FIG. 8G. FIG. 8B shows the state in which light spot 54 has come close to isolated magnetic domain 60. In this state, domain wall 61 has not reach yet front region 58, and the differential detection signal is at the base level as in FIG. 8A.
FIG. 8C shows the state in which domain wall 61 has just reached the boundary between front region 58 and the low temperature area. In this state, domain wall 61 of displacement layer 33 in front region 58 moves to a higher temperature region, and dotted line-shadowed portion 63 comes to be magnetized upward. The differential detection signal comes to the high level as shown in FIG. 8G. FIG. 8D shows the state in which domain wall 62 of the opposite side has reached the boundary between front region 58 and the low temperature area. In this state, domain wall 62 of displacement layer 33 in front region 58 moves to a higher temperature region to be magnetized downward, the differential detection signal returning to the base level as shown in FIG. 8G.
FIG. 8E shows the state in which light spot 54 proceeds further and domain wall 61 has reached the boundary between rear region 59 and the low temperature area. In this state, domain wall 61 of displacement layer 33 in rear region 58 moves to a higher temperature region, and dotted line-shadowed portion 64 comes to be magnetized upward. The differential detection signal comes to the middle level as shown in FIG. 8G. The signal level is lower than that of front region 58 because the center of the high temperature region is behind the center of light spot 54. FIG. 8F shows the state in which domain wall 62 of the opposite side has reached the boundary between rear region 59 and the low temperature area. In this state, domain wall 62 of displacement layer 33 in rear region 59 moves to a higher temperature region to change the magnetization direction downward, with the differential detection signal returning to the base level as shown in FIG. 8G.
In a one-beam method of domain wall displacement reproduction as described above, the domain walls are displaced in both the front region and the rear region to give two pulses for one isolated magnetic domain. In practical signals, since the magnetic domains are recorded arbitrarily, the displacements of the domain walls in the front region and the rear region contribute complicatedly to the differential detection signal, and such signals as a result of the complicate contributions cannot readily be separated.
The inventors of the present invention already applied for a patent on one method for solving the above problem (Japanese Patent Application No. 9-235885). This method is described briefly below by reference to FIGS. 7A and 7B and FIGS. 8A to 8G. As known by the figures, an exchange-coupled domain wall is delivered from the low temperature area to front region 58. On the other hand, at rear region 59, the domain wall is re-transferred at the boundary between the region and the low temperature area, and the domain wall is driven out toward the low temperature area. In this method, by application of a magnetic field, the re-transfer position is shifted from rear region 59 to the low temperature area to prevent movement of the domain wall within rear region 59.
In the above-mentioned method of applying a constant magnetic field, however, the accuracy of the movement of the domain wall (jitter) varies depending on the magnetization direction of the expanding magnetic domain. In the expansion of the magnetic domain magnetized in the same direction as the applied magnetic field, the external magnetic field serves to help the expansion to improve the accuracy of the movement, resulting in the decrease of the jitter. On the other hand, in expansion of the magnetic domain magnetized in the reverse direction to the applied magnetic field, the external magnetic field serves to retard the expansion to lower the accuracy of the movement, resulting in the increase of the jitter. Accordingly, in the pit-edge recording system for recording on a memory layer, as shown in FIGS. 3C to 3E, the jitter is satisfactory at the one side of the edge of a recorded magnetic domain, but the jitter at the other side of the edge is unsatisfactory, resulting in a lower linear density of the recording.